Testing requirements for
stormwater monitoring

There has been much discussion lately about the need for testing and verification of stormwater Best Management Practices (BMPs). Numerous testing and verification programs have evolved in recent years throughout North America. These include Environmental Technology Verification, Environmental Technology Evaluation Center, and Strategic Envirotechnology Partnership. These programs are a direct result of the proliferation of new environmental technologies which naturally raises questions regarding their performance. This has, in turn, led to greater scrutiny of stormwater monitoring in general.

Testing and verification programs are necessary to properly understand the function of innovative technologies.

The issues being debated include the validity of field testing, laboratory testing, and the need for "independent" field monitoring.

Field monitoring has drawbacks. Some automatic samplers are less than ideal since they may be prone to failure or may not take representative samples due to equipment limitations and pollutant characteristics (oil coating, composition alterations). Manual sampling is expensive, requires experience and still requires an estimation of flowrate. Strecker (1998) noted a study in Wisconsin which determined that the error in flow measurements using 23 different techniques was easily in the order of ±25% over a range of 50 storms (Waschbusch and Owens, 1998) compared to a calibrated Palmer-Bowlus flume. The error for individual storms was even greater.

Although field monitoring is problematic, it is a necessary part of the evaluation process. The alternative would be to accept laboratory testing or engineering calculations/computer modeling alone. Both of these alternatives, while useful, provide limited data upon which to make a decision.

Laboratory testing provides initial performance results under controlled conditions. Actual BMP installations operate under anything but controlled conditions. Accordingly, field performance varies based on numerous factors including climate, site conditions, land use, water temperature, interaction of pollutants, particle size distribution and variability of particle size distribution with rainfall intensity and flocculation. None of these can be properly accounted for in laboratory testing.

Similarly, stormwater related engineering calculations and computer modeling are based on numerous assumptions that over-simplify the pollutant removal process (e.g. discrete particle settling). Computer modeling is extremely useful in assessing the hydrological factors for design and provides planning level performance estimates for BMPs. These performance estimates, however, are limited by the underlying assumptions in each model/calculation.

Field monitoring can be used to augment laboratory testing and engineering calculations. If the field monitoring indicates better or worse performance than laboratory testing and theoretical calculations, then this data must be given considerable weight because it represents "real world" conditions. There are many site-specific variables which will affect the results of a single field study. Therefore, numerous field studies must be conducted under different climate, land usage, and pollutant loading conditions to determine the full range of BMP performance.

This underscores a prevalent problem with the current way in which field tests are summarized. Typically, BMP performance is provided in the form of a summary number (e.g. % Total Suspended Solids removal) without any supporting information regarding land use, pollutant loading, and site conditions. These factors directly influence the performance of a BMP and must be taken into account when reporting the effectiveness. These factors must also be considered when determining the usefulness of a BMP for water quality enhancement and when comparing different BMP technologies. In a sense, it is important to normalize the results based on site-specific conditions in order to compare apples to apples.

Another issue that is commonly debated is the need for independent monitoring. Independent monitoring refers to third party monitoring that is not paid for directly by the manufacturer or proponent of the BMP.

Many proponents of proprietary BMPs choose consultants to perform field monitoring to determine performance estimates. Consultant monitoring that is paid for by the proponent is seen to be less credible than agency testing. While it cannot be contested that there are good consultants and bad consultants, this attitude perpetuates a negative opinion about engineering consultants in general. A preferable solution would be to accredit consultants to perform field monitoring and provide a list of approved consultants who may be retained for proponent testing.

It is apparent that there are many pitfalls associated with performance testing and the reporting of performance results. Accordingly, an independent evaluation of a technology must be conducted, that considers all relevant data concerning that technology (field monitoring, laboratory testing, computer modeling), in order to provide a fair assessment of performance.

This four-step process (laboratory, theoretical/computer, field monitoring, independent evaluations), provides designers and regulators alike with a level of confidence to appropriately apply a given technology for stormwater quality.